BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:America/Los_Angeles BEGIN:DAYLIGHT DTSTART:20250309T030000 TZOFFSETFROM:-0800 TZOFFSETTO:-0700 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:PDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20251102T010000 TZOFFSETFROM:-0700 TZOFFSETTO:-0800 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:PST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20250604T030849Z UID:3B5E5007-0556-4DF0-BB1C-558D2611B06A DTSTART;TZID=America/Los_Angeles:20250603T190000 DTEND;TZID=America/Los_Angeles:20250603T200000 DESCRIPTION:Generative Adversarial Networks (GANs) implement Machine Learni ng (ML) algorithms that can address competitive resource allocation proble ms together with detection and mitigation of anomalous behavior. In this t alk\, we discuss their use in next-generation (NextG) communications withi n the context of cognitive networks to address i) spectrum sharing\, ii) d etecting anomalies\, and iii) mitigating security attacks. GANs have the f ollowing advantages. First\, they can learn and synthesize field data\, wh ich can be costly\, time consuming\, and non-repeatable. Second\, they ena ble pre-training classifiers by using semi-supervised data. Third\, they f acilitate increased resolution. Fourth\, they enable recovering corrupted bits in the spectrum. The talk will provide basics of GANs\, a comparative discussion on different kinds of GANs\, performance measures for GANs in computer vision and image processing as well as wireless applications\, se veral datasets for wireless applications\, performance measures for genera l classifiers\, a survey of the literature on GANs for i)–iii) above\, s ome simulation results\, and future research directions. In the spectrum s haring problem\, connections to cognitive wireless networks are establishe d. Simulation results show that a particular GAN implementation is better than a convolutional auto encoder for an outlier detection problem in spec trum sensing.\n\nSpeaker(s): Ender Ayanoglu\, \n\nVirtual: https://events. vtools.ieee.org/m/466461 LOCATION:Virtual: https://events.vtools.ieee.org/m/466461 ORGANIZER:isayan@ieee.org SEQUENCE:8 SUMMARY:Machine Learning in NextG Networks via Generative Adversarial Netwo rks URL;VALUE=URI:https://events.vtools.ieee.org/m/466461 X-ALT-DESC:Description: <br /><p class="MsoNormal">Generative Adversarial N etworks (GANs) implement Machine Learning (ML) algorithms that can address competitive resource allocation problems together with detection and miti gation of anomalous behavior. In this talk\, we discuss their use in next- generation (NextG) communications within the context of cognitive networks to address i) spectrum sharing\, ii) detecting anomalies\, and iii) mitig ating security attacks. GANs have the following advantages. First\, they c an learn and synthesize field data\, which can be costly\, time consuming\ , and non-repeatable. Second\, they enable pre-training classifiers by usi ng semi-supervised data. Third\, they facilitate increased resolution. Fou rth\, they enable recovering corrupted bits in the spectrum. The talk will provide basics of GANs\, a comparative discussion on different kinds of G ANs\, performance measures for GANs in computer vision and image processin g as well as wireless applications\, several datasets for wireless applica tions\, performance measures for general classifiers\, a survey of the lit erature on GANs for i)&ndash\;iii) above\, some simulation results\, and f uture research directions. In the spectrum sharing problem\, connections t o cognitive wireless networks are established. Simulation results show tha t a particular GAN implementation is better than a convolutional auto enco der for an outlier detection problem in spectrum sensing.</p>\n<p class="M soNormal">&nbsp\;</p> END:VEVENT END:VCALENDAR